Detection of Enzymes, Viruses, and Bacteria Using Glucose Meters

Subscriber access provided by University of South Dakota

Article

Rapid detection of enzymes, viruses and bacteria using glucose meters Amrita Das, Xikai Cui, Vasanta Chivukula, and Suri S. Iyer Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b02960 • Publication Date (Web): 07 Sep 2018 Downloaded from http://pubs.acs.org on September 8, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Rapid detection of enzymes, viruses and bacteria using glucose meters Amrita Das1, Xikai Cui1, Vasanta Chivukula2, and Suri S. Iyer 1,*

1 788

Petit Science Center, Department of Chemistry, Centre for Diagnostics and Therapeutics, Georgia State University, Atlanta, GA-30302.

2

Atlanta Metropolitan State College, 1630 Metropolitan Parkway, Atlanta, GA 30310.

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract: We have developed innovative assays that can detect enzymes rapidly. Paracetamol or catechol bearing compounds, when exposed to their respective enzymes released paracetamol or catechol, which can be detected using a standard glucose meter. This approach was used to detect a number of diverse analytes that include enzymes such as β-galactosidase, α-mannosidase and pathogens such as influenza viruses, Streptococcus Pneumoniae and E.coli rapidly. The limit of detection for all analytes was extremely low and clinically relevant for influenza viruses. We also demonstrate that glucose oxidase or glucose dehydrogenase is not required since the paracetamol gets oxidized directly on the electrode surface.

This indicates that test strips without glucose oxidase or

dehydrogenase can be used and we can detect analytes in the presence of high levels of background glucose. We demonstrate this unique nature of the assay to detect paracetamol in simulated urine and sheep blood without background interference of intrinsic glucose, indicating that glucose meters can be used to detect non-glucose analytes without background glucose interference.

Keywords: electrochemical detection, glucose meter, paracetamol, glycosidase, influenza, E. coli, assured diagnostics, point of care detection.


Page 2 of 29

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction: A number of clinically relevant enzymes provide a snapshot of the human condition. For example, blood tests to ascertain liver enzyme activity is part of regular blood tests to determine the health of a patient.

1-3

The standard method for diagnosis proceeds from

sample collection, shipping to a centralized laboratory and enzyme assays using fluorescent or chemiluminescent substrates. These methods, while clearly valuable, are time consuming and expensive. Translating biochemical assays performed in resource rich clinical or microbiological or biochemical laboratories by trained personnel to resource poor environments like primary clinics, remote locations and/or home by patients could arrest the spread of an emerging disease.4 For example, rapid detection of influenza results in early administration of antiviral drugs, which leads to faster recovery.5,6 Similarly, detection of environmentally harmful pathogens in food before the products are shipped to consumers is more valuable than expensive recalls of contaminated food. 7,8 The World Health Organization (WHO) has coined an acronym, ASSURED, for the requirements of Point of care (POC) diagnostic tests. ASSURED stands for Affordable, Selective, Sensitive, User-friendly, Rapid and Robust, Equipment free and Deliverable to end-users.

9

ASSURED diagnostics are critical for wide applicability of

diagnostic tests, especially for testing at remote locations and home use. Hundreds of biosensors have been designed to detect different analytes, but very few meet all of the ASSURED criteria as it is a very high bar to meet and often, there are trade-offs. However, if a diagnostic meets the ASSURED criteria, the impact can be transformative. For example, devices like the glucose monitor, which meets the ASSURED criteria have assisted millions of type 1 and 2 diabetic patients take more control of their blood glucose levels, thereby decreasing morbidity and healthcare costs significantly. 10,11 We and others have been promoting the use of glucose meters to detect non-glucose analytes. Liu and coworkers have used glucose meters to detect several analytes that include cocaine, adenosine, etc.

12,13

Their approach was to design assays that release

invertase upon target binding, which converts sucrose to glucose. (Figure 1A) On the other hand, we and others have developed substrates that release glucose when exposed to an enzyme.14-17 (Figure 1B) In both cases, the released glucose is oxidized by glucose oxidase or glucose hydrogenase, releasing electrons that can be recorded by a standard



glucose meter. Thus, the concentration of the analyte such as cocaine or the activity of the enzyme such as influenza neuraminidase can be correlated to the pathogen concentration. Using this approach, we demonstrated the detection of influenza viruses and measured drug susceptibility for influenza viruses as the FDA approved inhibitors, Relenza and Tamiflu are enzyme inhibitors. Glucose is not released when the inhibitors are added to the sample containing susceptible strains. All these assays use glucose meter for the detection of glucose or galactose. Galactose is detected if the dehydrogenase enzyme is impregnated on the test strips. 18,19 Taken together, a glucose meter can be repurposed to detect analytes other than blood glucose. This strategy is expected to have a significant impact in the development of point of care biosensors for a variety of ailments as millions of people have adopted glucose meters and there will be minimal training involved in using the same device for a new purpose. However, one potential concern in the use of the glucose meters and test strips for detection of analytes is the problem of background glucose in blood samples. In our previous work on influenza viruses, we used nasal swabs which is not as complex as blood and does not contain glucose unless the patient is suffering from an underlying condition. Also, we compared the glucose (or galactose level) before and after adding the substrate; the difference between these two readings allowed us to accurately determine the presence of influenza virus.14 While this strategy of measuring before and after addition of the substrate works well, background glucose levels can be an issue if the initial glucose reading is high especially when measuring blood samples from diabetic patients. Glucose meters are typically calibrated to measure the normal range of blood present in humans (70 -120 mg/dl) and anything lower or higher beyond this range is met with an error or “low” or “high” message.20 An alternate method that will overcome the problem of background glucose will allow us to adapt this strategy to detect analytes other than glucose in blood and other matrices. Here, we report novel assays, wherein paracetamol or catechol bearing compounds, when exposed to their respective enzymes, release paracetamol or catechol, which can be detected using standard glucose test strips and glucose meters. Paracetamol/catechol oxidation is a two electron and two proton process that has been described previously.21,22 We also demonstrate that strips without glucose oxidase or dehydrogenase can be used to


Page 4 of 29

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


detect paracetamol, suggesting that this method could be used to develop assays that can detect enzymes, viruses and bacteria in the presence of high levels of background glucose. Experimental section: I: Chemical Synthesis. General: Glycosylation reactions were performed under argon with solvents dried using a solvent purification system (Innovative Technology Inc., Amesbury, MA, USA). Other chemical reagents were of analytical grade, used as supplied, without further purification unless indicated. The acidic ion exchange resin used was Amberlite® IR-120 (H+) resin. Analytical thin layer chromatography (TLC) was performed on silica gel 230-400 mesh (Silicycle, Quebec City, Canada). Plates were visualized under UV light, and/or by staining with acidic CeH8Mo3N2O12, followed by heating. 1H and 13C NMR spectra were recorded on Bruker 400 MHz spectrometer. Chemical shifts are reported in δ (ppm) units using 13C and residual 1 H signals from deuterated solvents as references. Spectra were analyzed with MNova® (Mestrelab Research, Escondido, CA, USA). Electrospray ionization mass spectra were recorded on a Micromass QT 2 (Waters) and data were analyzed with MassLynx® 4.0 (Waters, Milford, MA, USA) software. Reported yields refer to spectroscopically and chromatographically pure compounds that were dried under high vacuum (10–2 mbar) before analytical characterization, unless otherwise specified. GP and MP were synthesized as described previously.23,24 Abbreviations: N, N Dimethyl formamide, DMF; Ethyl acetate, EA; Dichloromethane, DCM; Thin layer chromatography, TLC; Methanol, MeOH; Ethanol, EtOH; Triethylamine,

NEt3;

Tetrabutylammonium

hydrogensulfate,

TBAHSO4;

tert-

Butyldimethylsilyl chloride, TBSCl; 4-Dimethylaminepyridine, DMAP; Sulphuric acid, H2SO4; Trifluoromethanesulfonic acid, TMSOTf; Sodium methoxide, NaOMe; Sodium hydroxide, NaOH; Sodium carbonate, Na2CO3; Hydrochloric acid, HCl; Glucose oxidase, GOD. Methyl [4-N-acetamidophenyl-(5-acetamido-8,9-di-O-acetyl-3,5-dideoxy-Dglycero-α-D-galacto-2-ulopyranosyl)] onate (2):



To a solution of 1 (0.30 g, 0.64 mmol) in EA, paracetamol (0.12 g, 0.77 mmol) was added. TBAHSO4 (0.33 g, 0.96 mmol) was dissolved in 1M solution of Na2CO3 and was added dropwise. The biphasic reaction was stirred vigorously for 3 h. The reaction was monitored with TLC. On completion of the reaction, it was diluted with 20 ml of EA; washed with saturated Na2CO3 solution. (X2) The organic layers were collected and dried over Na2SO4 and concentrated under reduced pressure. The residue was subjected to flash silica gel column chromatography eluting with EA, to afford 2 (0.25 g, 63%) as white solid. 1H

NMR (400 MHz, CDCl3): δ 8.12 (s, 1H), 7.40 (d, J = 8Hz, 2H), 6.97 (d, J = 8Hz, 2H),

5.92 (d, J = 8Hz, 1H), 5.34-5.34 (m, 2H), 4.93-4.87 (m, 2H), 4.36-4.28 (m, 2H), 4.15-4.05 (m, 2H), 3.59 (s, 3H), 2.67 (dd, J = 20Hz, 8Hz, 1H), 2.10 (s, 3H), 2.09 (s, 3H), 2.07 (s, 3H), 2.01 (s, 3H), 1.99 (s, 3H), 1.85 (s, 3H). 13C NMR (100 MHz, CDCl3): δ 170.8, 170.5, 170.2, 168.6,167.9, 149.8, 134.5, 120.8, 100.2, 73.2, 69.3,68.9, 67.4, 62.0, 52.9, 49.2, 38.0, 24.3, 23.1, 21.0. ESI-MS: [M+Na]+: Calcd for C28H36N2O14Na : 647.2167; obtained 647.1951. 4-N-acetamidophenyl-(5-acetamido-3,5-dideoxy-D-glycero-α-D-galacto-2ulopyranoside)onic acid: (SP)

Compound 2 (0.05 g, 0.080 mmol) was dissolved in 1 ml of MeOH and treated with 30% NaOMe in MeOH ((100 µL) and stirred at rt for 1 h. The solution was neutralized with Amberlite IR 120(H+) resin, filtered and concentrated under reduced pressure. The residue was treated with aqueous 0.05 N NaOH and stirred at rt for 2 h. The reaction was monitored by TLC, the mixture was neutralized with Amberlite IR 120(H+) resin, filtered and concentrated under reduced pressure and subjected to P2 column to obtain pure compound SP (0.034 g, 94%). 1H NMR (400 MHz, D2O): δ 7.24 (d, J = 12 Hz, 2H),


Page 6 of 29

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


7.07 (d, J = 12 Hz, 2H), 3.83-3.75 (m, 6H), 3.71-3.65 (m, 2H), 3.57-3.49 (m, 4H), 2.79 (dd, J = 20 Hz, 8 Hz, 1H), 2.65 (s, 3H), 1.95 (s, 3H), 1.85 (t, J = 16 Hz, 1H). 13C NMR (100 MHz, D2O): δ 175.0, 172.9, 171.8, 150.5, 124.6, 123.2, 122.0, 115.7, 101.6, 73.5, 71.3, 68.1, 67.6, 62.8, 51.6, 40.1, 22.6, 22.0. ESI-MS m/z: [M-H]-: Calcd for C19H25N2O10: 441.1587; obtained 441.1443. Methyl [4-N-acetamidophenyl-(5-acetamido-8,9-di-O-acetyl- 3,5-dideoxy4,7-di-O-methyl D-glycero-α-D-galacto-2-ulopyranosyl)] onate (2a):

To a solution of 1a (0.40 g, 0.88 mmol) in EA, paracetamol (0.16 g, 1.1 mmol) was added. TBAHSO4 (0.45 g, 1.3 mmol) was dissolved in 1M solution of aqueous Na2CO3 and was added dropwise to the above stirring solution. The biphasic reaction was stirred vigorously for 3 h. The reaction was monitored with TLC. On completion of the reaction, it was diluted with 20 ml of EA; washed with saturated Na2CO3 solution. (X2) The organic layers were collected and dried over Na2SO4 and concentrated under reduced pressure. The residue was subjected to flash silica gel column chromatography eluting with EA, to afford 2a (0.29 g, 58%) as white solid. 1H NMR (400 MHz, CDCl3): δ 7.57 (s, 1H), 7.37 (d, J = 8Hz, 2H), 7.01 (d, J = 8 Hz, 2H), 5.56-5.54 (m, 1H), 5.36-5.26 (m, 2H), 4.62-4.59 (m, 1H), 4.25-4.09 (m, 4H), 3.63 (s, 3H), 3.52 (s, 3H), 3.34 (s, 3H), 2.81 (dd, J = 16 Hz, 8 Hz, 1H), 2.14 (s, 3H), 2.10 (s, 3H), 2.08 (s, 3H), 2.05 (s, 3H), 1.95 (t, J= 12 Hz, 1H). 13C NMR (100 MHz, CDCl3) δ 170.7, 170.6, 170.2, 168.4, 150.2, 134.1, 121.1, 100.7, 73.7, 70.9, 63.1, 61.4, 56.4, 53.5, 50.3, 37.7, 24.8, 23.4, 21.1, 13.9. ESI-MS m/z: [M+K]+: Calcd for C26H36N2O12 K: 607.2268; obtained 607.2259. 4-N-acetamidophenyl-(5-acetamido-3,5-dideoxy-4,7-di-O-methyl D-glyceroα-D-galacto-2-ulopyranoside)onic acid: SP(OMe)2



Compound 2a (0.03 g, 0.05 mmol) was dissolved in 1 ml of MeOH and treated with 30% NaOMe in MeOH (100 µL) and stirred at rt for 1 h. The solution was neutralized with Amberlite IR 120(H+) resin, filtered and concentrated under reduced pressure. The residue was then treated with aqueous 0.05 N NaOH and stirred at rt for 2h. The reaction was monitored by TLC, the mixture was neutralized with Amberlite IR 120(H+) resin, filtered and concentrated under reduced pressure and subjected to P2 column to obtain pure compound 3a (0.022 g, 98%). 1H NMR (400 MHz, D2O): δ 7.25 (d, J = 8 Hz, 2H), 7.05 (d, J = 8 Hz, 2H), 3.93-3.80 (m, 2H), 3.79-3.57 (m, 2H), 3.46 (s, 2H), 3.45-3.38 (m, 2H), 3.36-3.34 (m, 1H), 3.33 (s, 3H), 3.33 (s, 3H), 3.28-3.26 (m, 1H), 2.91 (dd, J = 12.5, 4.4 Hz, 1H), 2.06 (s, 3H), 1.96 (s, 3H), 1.73 (t, J = 12 Hz, 1H) 13C NMR (100 MHz, D2O): δ 174.1, 172.9, 172.5, 171.0, 165.9, 151.1, 123.3, 122.1, 102.9, 78.3, 78.0, 73.1, 71.4, 69.6, 62.2, 60.2, 56.7, 49.8, 37.4, 22.6, 22.2. ESI-MS m/z: [M-H]-: Calc. for C21H29N2O10: 469.1900; obtained 469.1828. 2-benzyloxyphenyl 2,3,4,6-tetra-O-acetyl-β-D-galactopyranoside (4):

To the stirring solution of galactose imidate 3 (0.20g, 0.40 mmol) in DCM, 2-(benzyloxy) phenol (0.10g, 0.54 mmol) was added. TMSOTf (1 ml, 0.04 mmol solution in DCM) was added dropwise at 0 0C. The reaction was monitored using TLC. Et3N was added to quench the reaction and warmed to rt. The reaction mixture was diluted with DCM, washed with Na2CO3 (X2) and brine solution (X2). The organic layers were collected and dried over Na2SO4 and concentrated under reduced pressure. The residue was subjected to flash silica gel column chromatography eluting with hexane:EA (40:60) to afford 4 (0.15 g, 71%) as white solid. 1H NMR (400 MHz, CDCl3): δ 7.46-7.39 (m, 4H), 7.35-


Page 8 of 29

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


7.34 (m, 1H), 7.19-7.17 (m, 1H), 7.05-7.03 (m, 1H), 6.99-6.90 (m, 2H), 5.58-5.54 (m, 1H), 5.45 (d, J=4Hz, 1H), 5.11-5.08 (m, 2H), 5.01 (d, J=8Hz, 1H), 4.26-4.13 (m, 3H), 4.00-3.97 (m, 1H), 2.20 (s, 3H), 2.06 (s, 3H), 2.02 (s, 3H), 1.82 (s, 3H) 13C NMR (100 MHz, D2O): δ 136.9, 128.5, 128.0, 127.4, 121.4, 120.2, 114.9, 101.0, 92.4, 85.4, 72.0, 70.9, 68.7, 66.2, 61.2, 29.7, 29.4, 20.6, 20.4, 17.4, 13.4. ESI-MS m/z: [M+K]-: Calc for C27H30O11K: 569.1830; obtained 569.3121. 2-hydroxyphenyl β-D-galactopyranoside (GC):

4 (0.05 g, 0.09 mmol) was dissolved in 1 ml of MeOH and treated with a 30% solution of NaOMe in MeOH (70 µL) and stirred at rt for 1 h. The solution was neutralized with Amberlite® IR 120 (H+) resin, filtered and concentrated to dryness. The dried compound was treated with Pd(OH)2/C (0.01 g) in absolute EtOH and was stirred for 12h at rt under H2 at 1 atm. After completion of reaction, reaction mixture was filtered using celite pad, washed with EtOH and combined solvent was concentrate to dryness and subjected to P2 gel column to furnish GC (0.023 g, 92%).1H NMR (400 MHz, D2O): δ 7.12 (d, J=8 Hz, 1H), 6.98-6.94 (m, 1H), 6.91-6.83 (m, 2H), 4.94 (d, J=8Hz, 1H), 3.97 (s, 1H), 3.92 (d, J= 4Hz, 1H), 3.80-3.68 (m, 4H), 3.58-3.54 (m, 1H), 3.49-3.45 (m, 1H). 13C NMR (100 MHz, D2O): δ 144.9, 124.1, 120.6, 116.8, 101.8, 75.4, 72.5, 70.5, 68.5, 62.5, 60.7. ESIMS m/z: [M-2H]-: Calc for C12H14O7: 270.2530; obtained 270.1854.

II. Biological assays: Materials: β- galactosidase from Escherichia coli and α-Mannosidase from Canavalia ensiformis (Jack Bean) were purchased from Sigma Aldrich, St. Louis, MO, USA. H3N2 (A/Aichi/2/1968), H1N1 (A/Brisbane/59/2007) were obtained from Beiresources (NIAID). S. pneumoniae ATCC 6301 and E.coli B (ATCC 11303) were obtained from American Type Culture Collection (ATCC). Glucose meter One Touch Ultra 2 and



Unistrip-1 generic blood glucose strips were purchased from Amazon. Defibrinated sheep blood was purchased from ThermoFisher Scientific (Waltham, MA) and treated with hexokinase to remove residual glucose. Artificial urine without glucose was purchased from Carolina Biologicals (Burlington, NC). Influenza virus growth: H3N2 (A/Aichi/2/1968) and H1N1 (A/Brisbane/59/2007) were grown and characterized as described previously.14,17 E. coli bacterial growth: The E. coli B (ATCC 11303) samples were grown in LuriaBertani (LB) broth for 20 h at 370C, supplemented with a final concentration of 1 mM isopropyl β -D-thiogalactopyranoside (IPTG) for the induction of β-galactosidase enzyme. The cells were harvested by centrifuging them at 2500 rpm for 5 min and re-suspended in sterile PBS buffer. 1% Triton was added to the E.coli cells to lyse and release the β galactosidase enzyme. S. pneumoniae bacterial growth: S. pneumoniae (ATCC 6301) was inoculated in Brain Heart Infusion (BHI) broth which was augmented with 0.16g/L of sialic acid to enhance NA production by the bacteria. Bacterial concentration was determined as 1.2 X 106 CFU/ml using colony count method. 100 μL of bacterial solution was lysed using 0.01% SDS and 20μl of chloroform was added to it. The sample was then lysed by vortexing for 30 sec and incubating at a temperature of 28 0C for 5 min. Detection: For all measurements, since the range of glucose meters has been set for testing normal blood glucose levels, low concentrations of glucose or paracetamol are recorded as “error”. Since we are detecting low concentrations of paracetamol, a known concentration of paracetamol is added to all solutions to establish a baseline number, which is subtracted from the assay readings. Detection of enzymes or E. coli: 100 μL of different concentrations of β-galactosidase or α-mannosidase or lysed E.coli was mixed with 100 ul of substrate GP or GC (3 mM) and 100 µl of paracetamol (3 mM) in PBS or citrate buffer at 370C for 1 h. 2 µL of the


Page 10 of 29

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


solution was used to measure the release of paracetamol using the Onetouch meter and Unistrips. Detection of influenza virus or S. pneumoniae: 100 μL of different concentrations of lysed H3N2 (A/Aichi/2/1968), H1N1 (A/Brisbane/59/2007) or S. pneumoniae was mixed with 100 µL of substrate SP or SP(OMe)2 (3 mM) and 100 ul of paracetamol (3 mM) in PBS buffer at 370C for 1 h. 2 µL of the solution was used to measure the release of paracetamol from the substrate.

Additionally, Figures S1-S5 and pertinent NMR images are provided in the Supporting information. Results and discussion: Our first step was to demonstrate that paracetamol can be detected using a glucose meter. To this end, we used a Onetouch glucose meter and standard strips to generate a calibration curve. The calibration curve and linear range for paracetamol demonstrates that it is possible to detect paracetamol using a glucose meter. (Supporting information, Figure S1) Since we could detect paracetamol readily and rapidly using glucose meters, we generated several compounds that could release paracetamol when exposed to the enzyme or pathogen. (Figure 2) The synthesis of the compounds is depicted in Scheme 1 and 2. The synthesis of compound SP is described below. Briefly, sialic acid chloride 1, was reacted with paracetamol to yield the α derivative in reasonably good yield. The H-3 axial and H-3 equatorial protons resonated at 1.7 and 2.8 ppm in the 1H NMR, respectively and the C-2 carbon resonated at 100.0 ppm in the 13C NMR, indicating α isomer. Deprotection of the acetate and the ester protecting groups resulted in the desired compound SP in high yield. We note that, unlike syntheses that introduces glucose or galactose into the compounds, the syntheses of paracetamol bearing compounds is relatively straightforward. NMR spectra of all intermediates and final compounds are provided in the Supporting information. The first assay we developed was to detect enzymes such as β-galactosidase and αmannosidase. We choose these glycosidases as they are commonly used in several



Page 12 of 29

biochemical assays to indicate the presence of microorganisms. GP23 was introduced to different concentrations of β-galactosidase and after different times, a drop of the solution (µL) was used to measure the paracetamol concentration. When 100 U of the enzyme was used, paracetamol was released almost instantaneously and the reaction was complete in 15 min. (Figure 3A) We determined the limit to be 0.01 U of enzyme in 5 min; (Supporting information, Figure S2) however, given a longer incubation time, the limit of detection can be decreased further as this is an enzyme detection assay. The linear range of detection is also given in the Supporting information, Figure S2B. To demonstrate that we can detect β-galactosidase in different matrices using this assay system, we spiked GP in different matrices and performed similar experiments. As seen in Figure 3B, matrix effects are not an issue. A similar result was obtained with MP 24 and α-mannosidase. (Figure 3C) These experiments demonstrate that we can reliably detect enzymes using paracetamol releasing substrates. Building on these results, we performed experiments to detect Escherichia coli. Detection of E. coli in recreational and drinking water is typically done by culture based methods and the disadvantage of these methods is that it requires a microbiology laboratory, trained personnel and it is time consuming.

25-27

Colorimetric methods are

available, but those tests rely on a visual color change, which might be subject to human error, especially when turbid or murky water samples are being tested.

25,28

The assays

described here use an electronic output and therefore, are not subject to the same human errors. Since β-galactosidase is released by the bacteria, we introduced GP to different concentrations of E.coli and measured the response at 15 min and 30 min. (Figure 3D and Supporting information, Figure S3). We could detect 102 CFU/ml in 15 min; however, a longer time of 1 h is required to obtain a better linear range because it is an enzyme based detection method. The next set of experiments were designed to demonstrate broad scope and to detect and differentiate influenza viruses from Streptococcus Pneumoniae. Briefly, influenza virus is a deadly respiratory pathogen that spreads rapidly. Early and accurate detection is important to reduce the disease burden. 29-31 Of particular importance is to differentiate between influenza and S. pneumoniae as this bacterium has been demonstrated to result in severe secondary illness after the initial attack by influenza viruses. 32-35 Previously, we


Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


reported the rapid and specific electrochemical detection and differentiation of influenza viruses in 15 min after sample processing using glucose and galactose bearing substrates. 14

Here, we used paracetamol bearing substrates to rapidly and specifically detect influenza viruses. When SP was exposed to a H3N2 strain of influenza virus (A/Aichi/2/1968) or S. Pneumonia (serotype 1, ATCC 6301), paracetamol was released as evidenced by the increase in the glucose meter readings, with S. Pneumonia producing a higher value. (Figure 4A). This is anticipated as the compound represents the natural substrate present on the termini of the glycocalyx of mammalian cells and all neuraminidases, viral, bacterial and human, cleave the natural substrate.

36-39

However,

when SP(OMe)2 was used, only influenza virus cleaves the substrate and not the bacteria (Figure 4B) because the methoxy groups interferes with the smaller binding pocket of the bacterial neuraminidase. 14,40-42 Therefore, this compound can be used to specifically detect influenza viruses. The limit of detection studies with two different strains of influenza viruses, a H1N1 (A/SolomonIslands/3/2006) and a H3N2 (A/Aichi/2/1968) strain indicate that we can detect 100 pfu/sample of these influenza strains rapidly. (Figure 4C, D) We also note that different strains could potentially be differentiated using this technique using continuous glucose meters; standard curves as a function of time for each strain could be developed and data from an unknown sample can be fitted to these curves to identify the specific strain. To demonstrate that background glucose does not affect the readings, we purchased commercial screen printed electrodes without any enzyme imprinted on them. Next, the strips were connected to a portable potentiostat and the current was measured when exposed to PBS buffer, which served as our negative control. As expected, there was no current generated as there is no analyte in solution. (Table 1) When exposed to 2 mM glucose solution, no current was generated as the enzyme is required to convert glucose to gluconolactone is not present. Next, the strips were exposed to 2 mM paracetamol solutions. We observed an increase in the current since paracetamol is directly oxidized on the surface of the electrodes. Finally, a mixture of 2mM glucose and 2 mM paracetamol was exposed to the strips. The increase in current was equal to the increase observed when only 2 mM paracetamol was used. The 2 mM glucose present in the solution that



Page 14 of 29

contained the mixture of 2 mM glucose and 2 mM paracetamol did not affect the reading, indicating that background glucose is not a problem. We repeated these experiments in different matrices, such as simulated urine and sheep blood that did not contain glucose to demonstrate there were no matrix effects. Briefly, either paracetamol or glucose or a mixture of paracetamol and glucose were spiked into these samples and the readings were taken using strips without any glucose oxidase. As seen in Table 1, there are no matrix effects and background glucose is not an issue when we are detecting paracetamol. The final set of experiments was designed to broaden the scope of the output, i.e., replacing paracetamol by catechol. This would allow more options for detection in the event that paracetamol bearing substrates are more complicated to synthesize or the sample contains high levels of paracetamol. First, we generated a standard curve for catechol. (Supporting information, Figure S4) Next, we synthesized a catechol bearing substrate GC (Figure 2) and exposed it to 100 U of β-galactosidase enzyme. (Supporting information, Figure S5) The enzyme was detected within 5 min, indicating that catechol or paracetamol bearing substrates could be used for the rapid detection of enzymes and pathogens using a personal glucose meter. Conclusions: We have been developing point of care diagnostics using glucose meters for the past several years. Our previous detection strategy involved the development of glucose or galactose based substrates. Exposure of these substrates to enzymes or pathogens released glucose or galactose, which was readily detected using a personal glucose meter. However, improvements to this strategy are required for several reasons. First, glucose is present in blood and in urine and other body fluids for diabetic and prediabetic patients. High background levels of glucose could cause a problem in the detection strategy. Second, detecting bacteria using glucose or galactose is problematic because the released glucose or galactose is used as a food source by E.coli, making it difficult to detect.43 In our investigations, we discovered that there are a few interfering chemical compounds that can perturb the glucose meter reading.44

One such compound is

acetaminophen or paracetamol, which can directly get oxidized at the electrode surface, which made us realize that we could incorporate paracetamol into the substrate. Cleavage


Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of the substrate will release paracetamol, which can be detected by the glucose meter. Therefore, we have introduced paracetamol or catechol as an alternate output molecule and demonstrated that this strategy can be used to detect enzymes, viruses and bacteria. An advantage is the ease of synthesis of these derivatives compared to more complex synthesis using glucose; the latter has five hydroxyl groups compared to one hydroxyl group present in paracetamol. A second advantage is that paracetamol and catechol have higher sensitivity than glucose, the limit of detection for paracetamol is 0.5 mM. In contrast, the limit of detection of glucose is 2 mM as measured using a Onetouch glucose meter. Thus, the detection limits of the enzyme in question can be decreased. A third advantage is the strips without glucose oxidase or dehydrogenase can be used to detect paracetamol. (Figure 5A, B, Table 1). Thus, background glucose or galactose will no longer be an issue if glucose oxidase or dehydrogenase is removed. The cost of producing the strips will also be reduced and the robustness of the strips will be increased as glucose oxidase or dehydrogenase is not required. One concern that could potentially arise is the presence of high levels of paracetamol in the blood or other body fluids. While paracetamol is not present in bodily fluids such as urine, blood or saliva, paracetamol is a pain medication. Fortunately, the peak plasma concentration reaches a maximum of ~ 20 micrograms, which is too low for detection by a glucose meter. 45-47 The concentration in peripheral blood is negligible and after a few hours, the drug is metabolized and excreted through urine as predominantly acetaminophen glucuronide.48 Thus, we are not overly concerned with background paracetamol levels and furthermore, catechol can be used instead of paracetamol. The assays described here compare very well with other established assays. For influenza viruses, the diagnostic tests available in the marketplace are are rapid influenza antigen detection tests (RIDTs), direct fluorescent-antibody assays (DFAs) and nucleic acid amplification tests (NAAT). The commercially available RIDTs for clinical samples provide a qualitative yes/no answer within 15-30 min. However, their sensitivity varies between 60-70% and can be lower, depending on the sample.49,50 DFAs are more sensitive compared to the RIDTs, but require around 2-3 h and skilled personnel to analyze the results. NAATs, e.g. Alere-i are highly sensitive and have a low detection limit of 101 to 103 pfu/ml and can be detected within 15-30 min but are very expensive.51 The electrochemical assay described here is inexpensive, user-friendly and can detect 100 pfu/sample of two strains of influenza rapidly. This assay also compares very well for



E.coli with established assays. Fluorescent based enzyme assays for E. coli focuses on the release of 4- Methylumbelliferyl-β-D-glucuronide (MUG) substrate. The hydrolysis of MUG releases fluorescent 4-methylumbelliferone (4-MU) and the intensity of the measured fluorescent signal is proportional to the amount of E. coli present in the sample. Using this method it is possible to detect 101-108 CFU/ml of E.coli in 20-120 min, depending on the bacterial count.52 A hydrogel based, E. coli detection system using a plunger-tube assembly, involving a colorimetric substrate can detect 105-106 CFU/ml of E.coli in 5 min.53 The assay reported here can detect 102 CFU/ml in 15 min. To summarize, we anticipate that this technique will be widely used for the rapid detection of a number of clinically relevant enzymes to report acute/chronic infections and human disorders as the technique meets the ASSURED criteria. References: (1) Limdi, J. K.; Hyde, G. M. Postgraduate medical journal 2003, 79, 307-312. (2) Hie-Won Hann , S. W., Ronald E. Myers, Richard S. Hann, Jinliang Xing, Bicui Chen, Hushan Yang. Plods ONE 7 2012, 10. (3) Kwo, P. Y.; Cohen, S. M.; Lim, J. K. Am J Gastroenterol 2017, 112, 18-35. (4) G Wu, M. Z. SciELO Public Health 2012, 90, 914–920. (5) Falsey, A. R.; Murata, Y.; Walsh, E. E. Archives of internal medicine 2007, 167, 354360. (6) Nelson, R. E.; Stockmann, C.; Hersh, A. L.; Pavia, A. T.; Korgenksi, K.; Daly, J. A.; Couturier, M. R.; Ampofo, K.; Thorell, E. A.; Doby, E. H.; Robison, J. A.; Blaschke, A. J. Pediatr Infect Dis J 2015, 34, 577-582. (7) Hoffmann, S.; Batz, M. B.; Morris, J. G., Jr. Journal of food protection 2012, 75, 12921302. (8) Cho, I. H.; Ku, S. International journal of molecular sciences 2017, 18. (9) Mabey, D.; Peeling, R. W.; Ustianowski, A.; Perkins, M. D. Nat Rev Micro 2004, 2, 231-240. (10) Schnell, O.; Erbach, M.; Wintergerst, E. J Diabetes Sci Technol 2013, 7, 904-912. (11) Diabetes Care 2016, 39 Suppl 1, S107-108. (12) Xiang, Y.; Lu, Y. Nature chemistry 2011, 3, 697-703.


Page 16 of 29

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(13) Lan, T.; Zhang, J.; Lu, Y. Biotechnol Adv 2016, 34, 331-341. (14) Cui, X.; Das, A.; Dhawane, A. N.; Sweeney, J.; Zhang, X.; Chivukula, V.; Iyer, S. S. Chemical Science 2017, 8, 3628 – 3634. (15) Mohapatra, H.; Phillips, S. T. Chem Commun (Camb) 2013, 49, 6134-6136. (16) Gurale, B. P.; Dhawane, A. N.; Cui, X.; Das, A.; Zhang, X.; Iyer, S. S. Analytical Chemistry 2016, 88, 4248-4253. (17) Zhang, X.; Dhawane, A. N.; Sweeney, J.; He, Y.; Vasireddi, M.; Iyer, S. S. Angew Chem Int Ed Engl 2015, 54, 5929-5932. (18) Dimeski, G.; Jones, B. W.; Tilley, V.; Greenslade, M. N.; Russell, A. W. Ann Clin Biochem 2010, 47, 358-365. (19) Ceriotti, F.; Kaczmarek, E.; Guerra, E.; Mastrantonio, F.; Lucarelli, F.; Valgimigli, F.; Mosca, A. Journal of Diabetes Science and Technology 2015, 9, 268-277. (20) Tonyushkina, K.; Nichols, J. H. Journal of diabetes science and technology (Online) 2009, 3, 971-980. (21) Engin, C.; Yilmaz, S.; Saglikoglu, G.; Yagmur, S.; Sadikoglu, M. Int J Electrochem Sc 2015, 10, 1916-1926. (22) Lin, Q. Q.; Li, Q.; Batchelor-McAuley, C.; Compton, R. G. J Phys Chem C 2015, 119, 1489-1495. (23) Mizuma, T.; Nagamine, Y.; Dobashi, A.; Awazu, S. Biochimica Et Biophysica ActaGeneral Subjects 1998, 1381, 340-346. (24) Han, Z. F.; Pinkner, J. S.; Ford, B.; Obermann, R.; Nolan, W.; Wildman, S. A.; Hobbs, D.; Ellenberger, T.; Cusumano, C. K.; Hultgren, S. J.; Janetka, J. W. Journal of Medicinal Chemistry 2010, 53, 4779-4792. (25) Fiksdal, L.; Tryland, I. Current opinion in biotechnology 2008, 19, 289-294. (26) Sicard, C.; Shek, N.; White, D.; Bowers, R. J.; Brown, R. S.; Brennan, J. D. Anal Bioanal Chem 2014, 406, 5395-5403. (27) Maheux, A. F.; Dion-Dupont, V.; Bouchard, S.; Bisson, M. A.; Bergeron, M. G.; Rodriguez, M. J. J Water Health 2015, 13, 340-352. (28) Deshmukh, R. A.; Joshi, K.; Bhand, S.; Roy, U. MicrobiologyOpen 2016, 5, 901-922. (29) Heron, M. Natl Vital Stat Rep 2016, 65, 1-95. (30) Short, K. R.; Habets, M. N.; Hermans, P. W. M.; Diavatopoulos, D. A. Future Microbiology 2012, 7, 609-624.



(31) New influenza A (H1N1) virus: WHO guidance on public health measures, J. Wkly Epidemiol Rec 2009, 84, 261-264. (32) Mitchell, A. M.; Mitchell, T. J. Clin Microbiol Infect 2010, 16, 411-418. (33) Krivan, H. C.; Roberts, D. D.; Ginsburg, V. Proc Natl Acad Sci U S A 1988, 85, 61576161. (34) Linder, T. E.; Daniels, R. L.; Lim, D. J.; DeMaria, T. F. Microb Pathog 1994, 16, 435441. (35) Tong, H. H.; Blue, L. E.; James, M. A.; DeMaria, T. F. Infect Immun 2000, 68, 921924. (36) Varghese, J. N.; Smith, P. W.; Sollis, S. L.; Blick, T. J.; Sahasrabudhe, A.; McKimmBreschkin, J. L.; Colman, P. M. Structure 1998, 6, 735-746. (37) Varghese, J. N.; McKimm-Breschkin, J. L.; Caldwell, J. B.; Kortt, A. A.; Colman, P. M. Proteins 1992, 14, 327-332. (38) Varghese, J. N.; Laver, W. G.; Colman, P. M. Nature 1983, 303, 35-40. (39) Gut, H.; Xu, G.; Taylor, G. L.; Walsh, M. A. J Mol Biol 2011, 409, 496-503. (40) Liav, A.; Hansjergen, J. A.; Achyuthan, K. E.; Shimasaki, C. D. Carbohydr Res 1999, 317, 198-203. (41) Shimasaki, C. D.; Achyuthan, K. E.; Hansjergen, J. A.; Appleman, J. R. Philos Trans R Soc Lond B Biol Sci 2001, 356, 1925-1931. (42) Achyuthan, K. E.; Pence, L. M.; Appleman, J. R.; Shimasaki, C. D. Luminescence 2003, 18, 131-139. (43) Chavali, R.; Gunda, N. S. K.; Naicker, S.; Mitra, S. K. Analytical Methods 2014, 6, 6223-6227. (44) Tang, Z.; Du, X.; Louie, R. F.; Kost, G. J. Am J Clin Pathol 2000, 113, 75-86. (45) Nielsen, J. C.; Bjerring, P.; Arendt-Nielsen, L.; Petterson, K.-J. European Journal of Clinical Pharmacology 1992, 42, 261-264. (46) Singla, N. K.; Parulan, C.; Samson, R.; Hutchinson, J.; Bushnell, R.; Beja, E. G.; Ang, R.; Royal, M. A. Pain Pract 2012, 12, 523-532. (47) Anderson, B. J.; Holford, N. H.; Woollard, G. A.; Chan, P. L. Br J Clin Pharmacol 1998, 46, 237-243. (48) Prescott, L. F. British Journal of Clinical Pharmacology 1980, 10, 291S-298S.


Page 18 of 29

Page 19 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(49) Beck, E.; Fan, J.; Hendrickson, K.; Kumar, S.; Shively, R.; Kramp, W.; Villanueva, J.; Jernigan, D.; Klimov, A.; Chen, L.-M.; Donis, R.; Williams, T.; Pirkle, J.; Barr, J. MMWR Morb Mortal Wkly Rep, Centers for Disease Control and Prevention, 2012, 61, 873-876. (50) Balish, A.; Warnes, C. M.; Wu, K.; Barnes, N.; Emery, S.; Berman, L.; Shu, B.; Lindstrom, S.; Xu, X.; Uyeki, T.; Shaw, M.; Klimov, A.; Villanueva, J. MMWR Wkly Rep 2009, 58, 826-829. (51) Nie, S.; Roth, R. B.; Stiles, J.; Mikhlina, A.; Lu, X.; Tang, Y.-W.; Babady, N. E. Journal of clinical microbiology 2014, 52, 3339-3344. (52) Hesari, N.; Alum, A.; Elzein, M.; Abbaszadegan, M. Enzyme Microb Technol 2016, 83, 22-28. (53) Gunda, N. S.; Chavali, R.; Mitra, S. K. Analyst 2016, 141, 2920-2929.

Acknowledgements: We are grateful to NIH-NIAID (RO1-A1089450) for funding. Conflict of Interest Disclosure: The authors declare no competing financial interest. Supporting information: Synthesis and characterization of all compounds and intermediates are provided in the Electronic Supplementary information.



1A

1B

Figure 1. Detection of non-glucose analytes using a glucose meter. A. DNA-invertase conjugates are immobilized onto magnetic beads. The DNA strands respond to the target of interest releasing DNA-invertase conjugates from the beads. The released invertase catalyzes the conversion of sucrose to glucose which can be quantified using a glucose meter. Because the concentration of DNA-invertase conjugate is proportional to the concentration of target, the glucose meter reading can be used to quantify the target concentration. B. Cleavage of substrate-glucose into substrate and glucose by the target enzyme. The released glucose and enzyme can be quantified using a glucose meter.


Page 20 of 29

Page 21 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. Structure of various glycoconjugates used as substrates for different enzymes. GP, Galactose-paracetamol cleaved by β-galactosidase; SP, Sialic acidparacetamol cleaved by Neuraminidase released by influenza virus and S. pneumoniae; MP, Mannose-paracetamol cleaved by α-mannosidase; SP(OMe)2, 4,7 methoxy sialic acid-paracetamol cleaved by influenza virus neuraminidase and GC, Galactose-catechol cleaved by β-galactosidase.



3A

3B 60 Sheep blood Artificial Urine PBS buffer

50

 GM reading

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 29

40 30 20 10 0 10 U

1U

0.1U 0.01U 0.001U

[-galactosidase] 3C

3D

Figure 3. Detection of enzymes and E.coli using Onetouch meters and Unistrips. A. Rapid detection of β-galactosidase: GP was incubated with 100 U of enzyme and monitored at different times. B. Rapid detection of β-galactosidase in various matrices: Various concentrations of β-galactosidase enzyme was added to artificial urine, defibrinated glucose-free sheep blood, and phosphate buffered saline (PBS) solutions and paracetamol was measured. C. Rapid detection of α-mannosidase: MP was incubated with 10 U of enzyme and monitored at different times. D. Rapid detection of E. coli and limit of detection studies: 1 mM of GP was incubated with different E. coli concentrations


Page 23 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and paracetamol was measured at 15 and 30 min. The y axis, Δ glucose meter reading, represents the difference in current between the negative control (no enzyme or E. coli) and the sample. All experiments were performed in triplicate (*p

Detection of Enzymes, Viruses, and Bacteria Using Glucose Meters

Recommend Documents